print "\n".join(['2013-11-%d' % day for day in range(8,11)])
1 Chase Manhattan Plaza
from IPython.core.display import display, Image
Image(filename='img/venue1.jpg')
Image(filename='img/venue2.jpg')
display(Image(filename='img/Borat_big_data.png'))
print "When dealing with big data avoid ETLs, bring code to data"
display(Image('img/Borat_80_20_rule.png'))
print "These days learning algorithms are easy to apply, most time is spent exploring the data and its features"
display(Image("img/heart.png", width=100))
display(Image("http://ipython.org/ipython-doc/rel-1.1.0/_images/ipynb_icon_128x128.png"))
print "Presentation media: "
print " iPython 20"
print " Keynote 1"
print " Google 1"
print " PowerPoint 0"
from sklearn import datasets
digits = datasets.load_digits()
print digits.DESCR[:52]
print digits.data.shape
print digits.target[0]
print digits.data[0]
for i in range(8):
print ["%2.0f" % d for d in digits.data[0][8*i:8*i+8]]
%matplotlib inline
import numpy as np
import matplotlib.pyplot as plt
fig, ax = plt.subplots(8, 12, subplot_kw={'xticks':[], 'yticks':[]})
for i in range(ax.size):
ax.flat[i].imshow(digits.data[i].reshape(8, 8), cmap=plt.cm.binary)
from sklearn.naive_bayes import GaussianNB
X = digits.data
y = digits.target
# Instantiate the estimator
clf = GaussianNB()
# Fit the estimator to the data, leaving out the last five samples
clf.fit(X[:-15], y[:-15])
# Use the model to predict the last five labels
y_pred = clf.predict(X[-15:])
print y_pred
print y[-15:]
print "sorry, no easy example to show :("
!~/anaconda/bin/conda info -e
!~/anaconda/bin/conda remove --name py3 --all --yes
!~/anaconda/bin/conda info -e
!~/anaconda/bin/conda create --name py3 python=3.3 pip --yes
!source ~/anaconda/bin/activate py3 && which python
!~/anaconda/bin/conda list --name py3
print "The Future: data gastronomification can use all 5 senses"
Image("https://github.com/csv/ddpy/raw/master/img/artichoke.jpg")
Image("img/datadog.png")